Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this...

10
International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705 www.rsisinternational.org Page 71 Fuzzy Logic Based Recommender System 1 Prof. Mehul Barot, 2 Dr. Kalpesh H. Wandra, 3 Dr. Samir B. Patel 1 Research Scholar, Computer Engineering Department, C.U.Shah University, Wadhwan City, India 2 Dean, C.U.Shah University, Wadhwan City, India 3 Assitant Professor, Pandit Deendayal Petroleum University, Gujarat, India Abstractwith current projections regarding the growth of Internet sales, online retailing raises many questions about how to market on the Net. A Recommender System (RS) is a composition of software tools that provides valuable piece of advice for items or services chosen by a user. Recommender systems are currently useful in both the research and in the commercial areas. Recommender systems are a means of personalizing a site and a solution to the customer’s information overload problem. Recommender Systems (RS) are software tools and techniques providing suggestions for items and/or services to be of use to a user. These systems are achieving widespread success in ecommerce applications now a days, with the advent of internet. This paper presents a categorical review of the field of recommender systems and describes the state-of- the-art of the recommendation methods that are usually classified into four categories: Content based Collaborative, Demographic and Hybrid systems. To build our recommender system we will use fuzzy logic and Markov chain algorithm. Keywords: Recommender System, Information Filtering, Prediction, Classification, User based, Item base, Fuzzy Logic. I. INTRODUCTION eb discovery applications like Stumble Upon, Reddit, Digg, Dice (Google Toolbar) etc to name a few are becoming increasingly popular on the World Wide Web. Information on the Internet grows rapidly and users should be directed to high quality Websites those are relevant to their personal interests. However, there is no way to Judge these web pages. Displaying quality content to users based on ratings or past Search results are not adequate. There‘s a lacking of powerful automated process combining human opinions with machine learning of personal preference. The goal of this project is to study recommendation engines and identify the shortcomings of traditional recommendation engines and to develop a web based recommendation engine by making use of user based collaborative filtering (CF) engine and combining context based results along with it using fuzzy logic and markov chain algorithm. The system makes use of numerical ratings of similar items between the active user and other users of the system to assess the similarity between users‘ profiles to predict recommendations of unseen items to active user. The system makes use of Pearson's correlation to evaluate the similarity between users. The results show that the system rests in its assumption that active users will always react constructively to items rated highly by similar users, shortage of ratings of some items, adapt quickly to change of user's interest, and identification of potential features of an item which could be of interest to the user. The System would benefit those users who have to scroll through pages of results to find relevant content. II. PROBLEM STATEMENT While studying recommender system, there were some hints at the problems that these companies have to overcome to build an effective recommender system. 1. Lack of Data Perhaps the biggest issue facing recommender systems is that they need a lot of data to effectively make recommendations. It‘s no coincidence that the companies most identified with having excellent recommendations are those with a lot of consumer user data: Google, Amazon, and Netflix. A good recommender system firstly needs item data (from a catalog or other form), then it must capture and analyze user data (behavioral events), and then the magic algorithm does its work. The more item and user data a recommender system has to work with, the stronger the chances of getting good recommendations. Figure 1: Data Gathering for Recommender System 2. Changing User Preferences Again suggested by Paul Edmunds, the issue here is that while today I have a particular intention when browsing e.g. Amazon tomorrow I might have a different intention. A classic example is that one day I will be browsing Amazon for new books for myself, but the next day I‘ll be on Amazon searching for a birthday present for my sister. On the topic of user preferences, recommender systems may also incorrectly label users. W

Transcript of Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this...

Page 1: Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this project is to study recommendation engines and identify the shortcomings of traditional

International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705

www.rsisinternational.org Page 71

Fuzzy Logic Based Recommender System

1Prof. Mehul Barot,

2Dr. Kalpesh H. Wandra,

3Dr. Samir B. Patel

1Research Scholar, Computer Engineering Department, C.U.Shah University, Wadhwan City, India

2Dean, C.U.Shah University, Wadhwan City, India

3Assitant Professor, Pandit Deendayal Petroleum University, Gujarat, India

Abstract— with current projections regarding the growth of

Internet sales, online retailing raises many questions about how

to market on the Net. A Recommender System (RS) is a

composition of software tools that provides valuable piece of

advice for items or services chosen by a user. Recommender

systems are currently useful in both the research and in the

commercial areas. Recommender systems are a means of

personalizing a site and a solution to the customer’s information

overload problem. Recommender Systems (RS) are software

tools and techniques providing suggestions for items and/or

services to be of use to a user. These systems are achieving

widespread success in ecommerce applications now a days, with

the advent of internet. This paper presents a categorical review

of the field of recommender systems and describes the state-of-

the-art of the recommendation methods that are usually

classified into four categories: Content based Collaborative,

Demographic and Hybrid systems. To build our recommender

system we will use fuzzy logic and Markov chain algorithm.

Keywords: Recommender System, Information Filtering,

Prediction, Classification, User based, Item base, Fuzzy Logic.

I. INTRODUCTION

eb discovery applications like Stumble Upon, Reddit,

Digg, Dice (Google Toolbar) etc to name a few are

becoming increasingly popular on the World Wide Web.

Information on the Internet grows rapidly and users should be

directed to high quality Websites those are relevant to their

personal interests. However, there is no way to Judge these

web pages. Displaying quality content to users based on

ratings or past Search results are not adequate. There‘s a

lacking of powerful automated process combining human

opinions with machine learning of personal preference.

The goal of this project is to study recommendation engines

and identify the shortcomings of traditional recommendation

engines and to develop a web based recommendation engine

by making use of user based collaborative filtering (CF)

engine and combining context based results along with it

using fuzzy logic and markov chain algorithm.

The system makes use of numerical ratings of similar items

between the active user and other users of the system to assess

the similarity between users‘ profiles to predict

recommendations of unseen items to active user. The system

makes use of Pearson's correlation to evaluate the similarity

between users.

The results show that the system rests in its assumption that

active users will always react constructively to items rated

highly by similar users, shortage of ratings of some items,

adapt quickly to change of user's interest, and identification of

potential features of an item which could be of interest to the

user. The System would benefit those users who have to scroll

through pages of results to find relevant content.

II. PROBLEM STATEMENT

While studying recommender system, there were some hints

at the problems that these companies have to overcome to

build an effective recommender system.

1. Lack of Data

Perhaps the biggest issue facing recommender systems is that

they need a lot of data to effectively make recommendations.

It‘s no coincidence that the companies most identified with

having excellent recommendations are those with a lot of

consumer user data: Google, Amazon, and Netflix. A good

recommender system firstly needs item data (from a catalog or

other form), then it must capture and analyze user data

(behavioral events), and then the magic algorithm does its

work. The more item and user data a recommender system has

to work with, the stronger the chances of getting good

recommendations.

Figure 1: Data Gathering for Recommender System

2. Changing User Preferences

Again suggested by Paul Edmunds, the issue here is that while

today I have a particular intention when browsing e.g.

Amazon – tomorrow I might have a different intention. A

classic example is that one day I will be browsing Amazon for

new books for myself, but the next day I‘ll be on Amazon

searching for a birthday present for my sister. On the topic of

user preferences, recommender systems may also incorrectly

label users.

W

Page 2: Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this project is to study recommendation engines and identify the shortcomings of traditional

International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705

www.rsisinternational.org Page 72

3. This Stuff is Complex!

Below slide illustrates that it takes a lot of variables to do

even the simplest recommendations. So far only a handful of

companies have really gotten recommendations to a high level

of user satisfaction – Amazon, Netflix (although of course

they are looking for a 10% improvement on their algorithm),

Google are some names that spring to mind. But for those

select few success stories, there are hundreds of other

websites and apps that are still struggling to find the magic

formula for recommending new products or content to their

users.

Figure 2: Variables needed for Recommendation

III. OBJECTIVE

Current recommender systems have a clear main objective: to

guide the user to useful/interesting objects. It is very

noticeable that this objective is composed of two different

tasks:

1. To generate suggestions to be accepted by the user.

2. To filter useful/interesting objects. The first task has

to do with the most external and interactive

behaviour that any recommender directly reveals to

the user. The second task is related to the known task

‗‗find good items‘‘, with a more internal and less

inter metrics published to date, it is difficult to

identify these two tasks together, as a whole

objective, on them. Moreover, a certain research bias

towards the second part of the objective could be

noticed, while frequently losing the first part.

IV. SCOPE OF RECCOMENDER SYSTEM

Recommender System are the software engines and

approaches for providing suggestion of products to the user

which might be most probably matched to the user‘s choice.

Usually the recommender system is a technology which filters

out the information to envision in case a particular user will

like a specific item; this is usually called as prediction

problem, or to identify N set of items that will be of certain

users interest called as Top N recommendation problem. From

past few years the use of recommender System is being

gradually increasing in various different applications, for

instance application for recommending books, CDs and other

products at different search engines like amazon.com ,

Netflix.com, ebay.com and so on. Even the Microsoft

suggests many additional software‘s to user, to fix the bugs

and so forth. When a user downloads some software, a list of

software is provided by the system. All the above examples

would be result of diverse service, but all of them are

categorized into a recommendation System, Identifying web-

pages that will be of interest, or even implying backup ways

of searching for information‘s.

V. LITERATURE SURVEY

In the last sixteen years, more than 200 research articles were

published about research-paper recommender systems. I found

that more than half of the recommendation approaches applied

content-based filtering (55%). Collaborative filtering was

applied by only 18% of the reviewed approaches, and graph-

based recommendations by 16%.The use of efficient and

accurate recommendation techniques is very important for a

system that will provide good and useful recommendation to

its users.

Table 1- Literature survey-1

Page 3: Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this project is to study recommendation engines and identify the shortcomings of traditional

International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705

www.rsisinternational.org Page 73

VI. RECCOMENDATION SYSTEM

Recommendation system is an information filtering technique,

which provides users with information, which he/she may be

interested in.

1. Classification of Recommendation Systems:

Most of the recommendation systems can be classified into

either User based collaborative filtering systems or Item based

collaborative filtering systems. In user based collaborative

filtering a social network of users sharing same rating patterns

is created. Then the most similar user is selected and a

recommendation is provided to the user based on an item

rated by most similar user. In item based collaborative

filtering relationship between different items is established

then making use of the active user's data and the relationship

between items a prediction is made for the active user. [25]

2. Methodologies

The proposed system makes use of Pearson‘s correlation to

implement User based collaborative filtering, and context,

Synonym Finder to implement Context based filtering

techniques to generate recommendations for the active user.

Following are the methodologies used/researched so far:

Taste:

Taste is a flexible, fast collaborative filtering engine for

Java. It takes the users' preferences for items and The

engine takes users' preferences for items ("tastes") and

recommends other similar items [25]

Vogoo:

Vogoo is a php based collaborative filtering and

recommendation library. It recommends items to users,

which matches their tastes. It calculates similarities

between users and creates communities based on them.

The figure below shows the results of using vogoo to

generate similar taste sharing users and recommendations

made by the most similar users [25]

Fuzzy Logic:

Here I tried to make use of fuzzy logic to calculate

similar users.

Following is the currently used approach:

User Request:

User makes a request for recommendation by clicking on

the recommendation menu. User is asked to provide

contextual information.

Server:

The information provided by the user is send to the

server. The server is composed on 2 sub engines: user

based collaborative filtering engine, and context based

engine. The server sends users request to both the sub

engines.

User based collaborative filtering engine: - calculates

similar users based on the numerical ratings of common

items rated by the active users and other users of the

system. The system achieves this by making user of the

Pearson‘s correlation

Pearson’s Correlation:

It is a way to find out similar users. The correlation is a way to

represent data sets on graph. Pearson‘s correlation is x-y axis

graph where we have a straight line known as the best fit as it

comes as close to all the items on the chart as possible. If two

users rated the books identically then this would result as a

straight line (diagonal) and would pass through every books

rated by the users. The resultant score is this case is 1. The

more the users disagree from each other the lower their

similarity score would be from 1. Pearson‘s Correlation helps

correct grade inflation. Suppose a user ‗A‘ tends to give high

scores than user ‗B‘ but both tend to like the book they rated.

The correlation could still give perfect score if the differences

between their scores are consistent.

Inaccurate queries: We have user typically domain specific

knowledge. And users don‘t include all potential Synonyms

and variations in the query, actually user have a problem but

aren‘t sure how to phrase.

VII. PROPOSED ARCHITECTURE

Figure 3: Proposed Architecture

Description:

1. User types in the URL for the system on a Web

Browser.

2. User logs into the system using his `userid`.

3. The user chooses from amongst the type 2 different

types of recommendation systems available.

4. If the user chose ‗Collaborative Filtering‘ option, the

system calculates similar users making use of

Table 2- Literature survey-2

Page 4: Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this project is to study recommendation engines and identify the shortcomings of traditional

International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705

www.rsisinternational.org Page 74

engineering algorithms, and then recommends items

to the users based on the most similar user.

5. If the user chose ‗Context based Filtering‘ option, the

system then makes use of the context information,

and Synonym Finder to make predictions.

VIII. PROPOSED METHOD

Fuzzy Set and Fuzzy Logic

Fuzzy set theory consists of mathematical approaches that are

flexible and well-suited to handle incomplete information, the

un-sharpness of classes of objects or situations, or the

gradualness of preference profiles. Fuzzy set theory and logic

provide a way to quantify the uncertainty due to vagueness

and imprecision. Membership functions, a building block of

fuzzy sets, have possibilistic interpretation, which assumes the

presence of a property and compares its strength in relation to

other members of the set. A fuzzy set A in X is characterized

by its membership, which is defined as: (x) : x X [0,1] µ A ,

where X is a domain space or universe of discourse.

Alternatively, A can be characterized by a set of pairs:

{ )) ……….. (1)

According to the context in which X is used and the concept

to be represented, the fuzzy membership function, (x µ A ),

can have different interpretations. As a degree of similarity, it

represents the proximity between different pieces of

information. For example, movie x in the fuzzy set of

"electronics" can be estimated by the degree of similarity. As

degree of preference, it represents the intensity of preference

in favor of x, or the feasibility of selecting x as a value of X.

For instance, a product rating of 4 out of 5 indicates the

degree of a user's satisfaction or liking with x based on certain

criteria.

Formalism of the Representation and Inference Methods

The proposed fuzzy theoretic content-based approach is based

on a user‘s previous feedbacks, and features of the new items

and features of the set of items for which the user has

provided feedback. The rationale of this method is users are

more likely to have interest in item like movie that is similar

to the items like movies they have experienced and liked. This

approach is useful for new item like movie with no or few

user ratings and purchase. It is solely based on one user‘s

previous interest expressed by ratings. The representation

scheme, inference engine consisting of recommendation

strategies and similarity measures, and the algorithm of the

proposed method are presented in this section.

Items Representation Using Fuzzy Set

For an item described with multiple attributes, more than one

attribute can be used for recommendation. Moreover, some

attributes can be multi-valued involving overlapping or not

mutually exclusive possible values. For example, products are

multi-categorized and multi-functional. These values of multi-

valued attributes in an item can be represented more

accurately with in a fuzzy set framework than with in a crisp

set framework. Let an item Ij (j = 1 … M) be defined in the

space of an attribute X ={x1, x2, x3, …. xL}, then Ij can take

multiple values such as x1, x2, …, and xL. If these values of

X can be sorted in the decreasing order of their presence in the

item Ij expressed by degrees of membership, then the

membership function of item Ij to value xk (k = 1 …L),

denoted byµ (I ) x j k , can be obtained heuristically. Hence, a

vector formed for Ij:

{ ) ) …….. (2)

µ (I ) xj k can be interpreted as the degree of similarity of Ij

to a hypothetical (or prototype) pure xk type of the item; or as

the degree of presence of value xk in item Ij.

Fuzzy Theoretic Similarity Measures

One of the most important issues in recommender systems

research is computing similarity between users, and between

objects (items, events, etc.). This in turns highly depends on

the appropriateness and accuracy of the methods of

representation. In fuzzy set and possibility framework,

similarity of users or items is computed based on the

membership functions of the fuzzy sets associated to the users

or items features. Similarity is studied and applied in

taxonomy, psychology and statistics. Similarity is subjective

and context dependent. The set-theoretic, proximity-based and

logic-based are the three classes of measures of similarity.

Based on the results of the study those measures that are

relevant for items recommendation application are adapted.

IX. ANALYSIS

In below table essential parameters are discussed below along

with their respective meanings and possible values i.e Yes: Y,

No: N, Not Discussed: ND.

Serial No Parameters

Meaning of

Parameter Justify Possible Values

1 Efficiency A level of performance that Yes, No, Not Discussed describes a process that uses the

lowest amount of inputs to create

the greatest amount of outputs in

minimum time & memory

2 Accuracy in terms of prediction Accuracy of algorithms should be Yes, No, Not Discussed Maintained. Error rate should be

Minimized.

Page 5: Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this project is to study recommendation engines and identify the shortcomings of traditional

International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705

www.rsisinternational.org Page 75

3 User stratification User specifications and needs Yes, No, Not Discussed must be satisfied

4 Automatable Methods describes are Yes, No, Not Discussed automatable which reduces

manual work

5 Robustness Specification are may not be Yes, No, Not Discussed covered and but appropriate

performance of a system

6 Integration Integration of system allows Yes, No, Not Discussed combination of 2 concepts such that system is capable of

producing better results.

7 Flexible Flexibility refers to designs that Yes, No, Not Discussed can adapt when external changes

occur.

8 Performance Backtracking or recovery process Yes, No, Not Discussed defines the performance of the

system

9 Satisfaction of the Recommendation Provider User should be satisfied with Yes, No, Not Discussed recommended results. Relevant

information should be provided

in order to win out user

satisfaction.

10 Diversity The concept of diversity Yes, No, Not Discussed encompasses acceptance and

unique.

11 Timing constraint Appropriate timing is associated Yes, No, Not Discussed with every algorithm

12 Effortless The quality of a system that Yes, No, Not Discussed makes the user to use it easily

13 Optimization Optimization is the process of Yes, No, Not Discussed modifying a system to make

some features of it work more

efficiently or use fewer resources

14 Error rate It measures the total number of Yes, No, Not Discussed incorrect predictions against the

total number of predictions

15 Precision It is defined where datasets are Yes, No, Not Discussed much unbalanced.

16 Recall It is the proportion of the number Yes, No, Not Discussed of data items that system selected

as the positive

17 F1-Score For optimization F1 score Yes, No, Not Discussed combines both recall and

precision with equal importance

into a one parameter.

18 Receiver Operating Characteristic (ROC) graph It is technique to organize, Yes, No, Not Discussed visualize, and select classifiers

that depend on their performance

in 2D space.

X. PRIOR AND RELATED WORK

As merchandisers gained the ability to record transaction data,

they started collecting and analyzing data about consumer

behavior. The term data mining is used to describe the

collection of analysis techniques used to infer rules from or

build models from large data sets. One of the best-

known examples of data mining in commerce is the discovery

of association rules—relationships between items that indicate

a relationship between the purchase of one item and the

purchase of another. These rules can help a merchandiser

arrange products so that, for example, consumer purchasing

ketchup sees relish nearby.

More sophisticated temporal data mining may suggest that a

consumer who buys a new charcoal grill today is likely to buy

a fire extinguisher in the next month. More generally, data

mining has two phases. In the learning phase, the data mining

system analyzes the data and builds a model of consumer

behavior (e.g., association rules). This phase is often very

time-consuming and may require the assistance of human

analysts. After the model is built, the system enters a use

Table-3 Evaluation parameters for product recommendations

Page 6: Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this project is to study recommendation engines and identify the shortcomings of traditional

International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705

www.rsisinternational.org Page 76

phase where the model can be rapidly and easily applied to

consumer situations. One of the challenges in implementing

data mining within organizations is creating the organizational

processes that successfully transfer the knowledge from the

learning phase into practice in the use phase. Automatic

recommender systems are machine learning systems

specialized to recommend products in commerce applications.

XI. POPOSED ALGORITHM

Fuzzy Preference Tree-Based Recommendation Approach

In this algorithm it intends to cover both the user‘s

intentionally expressed preference and their extensionally

expressed preference from the user items. It form the structure

based on two factors such as matching the corresponding the

parts and rating by prediction on user targeted item using user

preference aggregations. Next the two trees are mapped using

conceptual similarities between the corresponding parts of two

trees with fuzzy preference.

• Here the user‘s fuzzy preference tree is mentioned as

and the item tree

• The maximum conceptual similarity tree mapping

between and

• Then both trees are weighed equally and constructed

by merging into

• The tree operation is done and the merging is defined

by Next the function pr (), that takes the fuzzy

preference tree and similarity tree mapping as input

Then it works as follows:

Algorithm: Rating prediction algorithm. [19]

[ ] )

Input: Fuzzy preference tree node

Output: the predicted rating

1. ( [ ] )

2. [ ])

3. Return 0;

4. Else if [ ])

5. Let the preference value be {

6. Return ∑

7. Else if [ ])

8. Return ∑ [ ] ) [ ]

9. Else if [ ] [ ]) )

10. Return

∑ ( ) ∑ [ ]

[ ] )

Hybrid Filtering

Both content-based filtering and collaborative filtering have

their strengths and weaknesses. Three specific problems can

be distinguished for content-based filtering:

Content description. In some domains generating a

useful description of the content can be very difficult.

In domains where the items consist of music or video

for example a representation of the content is not

always possible with today‘s technology.

Over-specialization. A content-based filtering system

will not select items if the previous user behaviour

does not provide evidence for this. Additional

techniques have to be added to give the system the

capability to make suggestion outside the scope of

what the user has already shown interest in.

Subjective domain problem. Content-based filtering

techniques have difficulty in distinguishing between

subjective information such as points of views and

humour. [13]

A collaborative filtering system doesn‘t have these

shortcomings. Because there is no need for a description of

the items being recommended, the system can deal with any

kind of information. Furthermore, the system is able to

recommend items to the user which may have a very different

content from what the user has indicated to be interested in

before. Finally, because recommendations are based on the

opinions of others it is well suited for subjective domains like

art. However, collaborative filtering does introduce certain

problems of its own:

Early ratter problem. Collaborative filtering systems

cannot provide recommendations for new items since

there are no user ratings on which to base a

prediction. Even if users start rating the item it will

take some time before the item has received enough

ratings in order to make accurate recommendations.

Similarly, recommendations will also be inaccurate

for new users who have rated few items.

Scarsity problem. In many information domains the

existing number of items exceeds the amount a

person is able (and willing) to explore by far. This

makes it hard to find items that are rated by enough

people on which to base predictions.

Grey sheep. Groups of users are needed with

overlapping characteristics. Even if such groups

exist, individuals who do not consistently agree or

disagree with any group of people will receive

inaccurate recommendations.

A system that combines content-based filtering and

collaborative filtering could take advantage from both the

representation of the content as well as the similarities among

users. Although there are several ways in which to combine

the two techniques a distinction can be made between two

basis approaches. A hybrid approach combines the two types

of information while it is also possible to use the

recommendations of the two filtering techniques

independently. [13]

We used MovieLens10k dataset using hybrid filtering

technique and Pearson correlation similarity.

Page 7: Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this project is to study recommendation engines and identify the shortcomings of traditional

International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705

www.rsisinternational.org Page 77

XII. COMPARATIVE ANALYSIS OF ALGORITHAM

These algorithms show varied performance under different

training sets. The dimensionality-reduction algorithm requires

the highest runtime. The item-based algorithm works slowly

for the large number of products. The spreading-activation

and link-analysis algorithms require less number of iterations

to achieve acceptable recommendation quality. The generative

model algorithm is very efficient as it needs a small number of

hidden classes for quality recommendations. The spreading-

activation algorithm is especially fast. The link-analysis

algorithm usually performs the best

Collaborative Advantages Disadvantages

Filtering

Recommendation

Algorithm

User-based Simple to implement Suffer serious

Algorithm[27]. Scalability

problems

Due to sparsity

accuracy is low.

Item-based Better performance Slow for large

Algorithm

[27]. & quality than user number of based algorithm.

items.

Less computation.

Provides higher

efficiency.

Faster than user

based algorithm

Dimensionality- Simplifies the Requires the

reduction sparsity problem. highest runtime.

Algorithm. [27]

Generative-model Scalable Expensive due

Algorithm [27] Flexible model

building

One can lose

useful

information

due to reduction

models.

Spreading- Relaxes the sparsity Works only

activation & cold start when sufficient

Algorithm[27] Problem data is not

Fast as it computes

available

Recommendation

only for

target consumers

Link-analysis Better performance Works only

Algorithm [27]

than user when sparse

based & Item data is available

based algorithm.

Useful when sparse

data is available.

Table-4 Comparative Analysis of Collaborative Filtering

Recommendation Algorithm [27]

Recommendation Advantages Disadvantages/Limit

Strategy ations

Content Based User Limited content

Recommendation Independence analysis

Transparency Over-specialization

Collaborative Easy to create It totally depends on

Filtering and use human ratings

Explain- Sparsity( Insufficient

ability of the data)

Results Scalability

New data can Ignore the social

be added relationships.

Easily Cold start problem

More (Low Accuracy)

applicable in

the reality

Trust Based Avoid cold Difficult to develop a

Recommendations start problem. trust network

Alleviate the

sparsity

problem

Figure 4. Comparative analysis of algorithm

XIII. IMPLEMENTATION

There are various level of implementation took part to build

this system some important of them is listed below:

A. Dataset and Preprocessing.

The benchmark dataset from MovieLens at the University of

Minnesota (http://movielens.umn.edu), which has been widely

used in recommendation research, is employed in this study.

The dataset includes movie attributes, user ratings, and simple

user demographic information. The dataset consists of

100,000 ratings (1-5) from 943 users on 1682 movies; and

each user has rated at least 20 and at most 737 movies.

Genres in the MovieLens dataset are represented with binary

values, which do not reflect the true content of movies in the

genre space. Therefore, we use the proposed representation

scheme by incorporating information about movie genres

Table 5-Comparative Analysis of Recommendation Strategies [27]

Page 8: Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this project is to study recommendation engines and identify the shortcomings of traditional

International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705

www.rsisinternational.org Page 78

from the Internet Movie Database (imdb.com), which is a

large database consisting of comprehensive information about

past, present and upcoming movies.

B. Implement similarity model

After data cleaning process, actual implementation process

started. In this phase I build recommender system based on

user-user similarity and item-item similarity as well. I used

Pearson correlation similarity and used mahout for this

system.

Recommender systems for movies are designed and

developed to assess the effectiveness of the proposed

methods. The system works as follows:

a) For each user, it randomly splits the ratings dataset

into a training set and a test set.

b) Using the training set, it computes recommendation

confidence score for each item in the test set using

the different similarity measures and

recommendation strategies.

c) For each user, using the movies in the testing set, it

computes Top-N recommendations and the

recommendation accuracy– precision, recall, and F1–

measure. Moreover, computational time for learning

user preference and presenting the recommendation

are recorded.

d) Using different random selection of the movies into

testing and training sets, 10 different runs are

executed to avoid sensitivity to sampling bias, and

the average results are reported

C. Evaluation metrics

Accuracy is a commonly used metric for a recommender

system based on user tasks or goals. The accuracy metrics

includes predictive and recommendation accuracy measures.

Predictive accuracy measures such as mean absolute error,

mean square error and percentage of correct predictions are

found to be less appropriate when the user task is to find

‗good‘ items and when the granularity of true value is small

because predicting a 4 as 5 or a 3 as 2 makes no difference to

the user. Instead, recommendation accuracy metrics including

recall, precision and F1-measures are more appropriate.

Approximations to the true precision and recall are computed

using movies for which ratings are provided and held for

testing. This approach of measuring performance is widely

used in recommender systems research.

Precision measures the ratio of correct recommendations

being made. Recall reflects the coverage or hit rate of

recommendations.

F1-Measure = (2*precision*recall)/(precision + recall)

is a single metric that combines precision and recall. They are

defined as:

Precision = (no of movies in the TOP-N movies with

rating b4) / N

Recall = (no of movies in the TOP-N with rating) / M

Where M is total number of movies rated as interesting in

the test cases; and N is total number of movies

recommended.

XIV. RESULTS

Mean of recommendation accuracies by recommendation

strategy and similarity measure are presented in Figure 5. The

maximum mean precision, recall and F1 measure are around

53%, 27%, 36% for FLRS compared to the 55%, 38%, and

38% for FTM and 49%, 39% and 38% obtained for the CSM

(Crisp Set Similarity based) approach, respectively.

Figure: 5 Average/Mean recommendation accuracies by similarity measure

The Figure 6 and 7 indicates the proposed approach improves

the performance of user-based CF method. Based on

evaluated results our approach has more accurate results and

therefore our approach shows more qualified

recommendations compared to fig 8 which is results of user

based CF system. [25]

Figure: 6 Precision measures for top N recommendations

0

0.1

0.2

0.3

0.4

0.5

0.6

Precision Recall Fmeasure

Chart Title

FTM CSM FLRS

0

50

100

150

200

250

0 5 10 15

Pre

cisi

on

k

Page 9: Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this project is to study recommendation engines and identify the shortcomings of traditional

International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705

www.rsisinternational.org Page 79

Figure: 7 Recall measures for top N recommendations

Figure: 8 Precision and recall of user based CF.

The Figure 9 and 10 indicates the proposed approach

improves the performance of FARS method. Based on

evaluated results our approach has more accurate results and

therefore our approach shows more qualified

recommendations. [23]

Figure: 9 Precision measures for top N recommendations

Figure: 10 Recall measures for top N recommendations

Figure 11 and 12 indicates the proposed approach

improves the performance of FARS.

Figure: 11 Precision measures for N top recommendations

Figure: 12 Recall measures for N top recommendations

XV. CONCLUSION AND FUTURE WORK

Recommender systems are tools which provide a personalized

environment for the users of a web site by investigating their

navigational behavior in a period of time. In this paper a fuzzy

logic based recommender system (FLRS) was proposed. This

research develops a fuzzy logic methodology for

recommender systems. Using actual data on movies, the

results of the research integrating user and item features, and

using fuzzy and possibility theories as the foundation for

representing and reasoning about uncertainty contribute to the

0

50

100

150

200

250

0 5 10 15

Rec

all

k

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 5 10 15 20 25

pre

cisi

on

k

precision

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 5 10 15 20 25

reca

ll

k

Page 10: Fuzzy Logic Based Recommender System · Fuzzy Logic Based Recommender System ... The goal of this project is to study recommendation engines and identify the shortcomings of traditional

International Journal of Research and Scientific Innovation (IJRSI) | Volume IV, Issue VIS, June 2017 | ISSN 2321–2705

www.rsisinternational.org Page 80

effectiveness and efficiency of the proposed method for

recommender systems.

We performed experimental comparison of our proposed

method against well-known user based CF, FARS, FTM and

CSM approach.

Further studies are planned to extend this approach in several

directions. First, inclusion of additional attributes for movies

expected to improve the performance of the system. Second,

test the FTM approach with additional dynamic datasets and

domain applications to see the generalization of the results.

REFERENCES

[1]. Andomavicius, G. and A. Tuzhilin,‖ Toward the next generation

of recommender system: A survey of the state-of-the- art and

possible extensions‖. IEEE Trans, Knowledge Data Eng., 17: 734-749, 2005.

[2]. J. Canny. Collaborative filtering with privacy via factor analysis.

In Proc. of SIGIR ‘02, pages 238–245, Tampere, Finland, 2002. [3]. T. Hofmann. Collaborative filtering via gaussian probabilistic

latent semantic analysis. In Proc. of SIGIR ‘03, pages 259–266,

Toronto, Canada, 2003. [4]. J. D. M. Rennie and N. Srebro. Fast maximum margin matrix

factorization for collaborative prediction. In Proc. of ICML ‘05,

2005. [5]. R. Salakhutdinov and A. Mnih. Bayesian probabilistic matrix

factorization using markov chain monte carlo. In Proc. of ICML

‘08, 2008. [6]. D. Agarwal and B.-C. Chen. flda: matrix factorization through

latent dirichlet allocation. In Proc. of WSDM ‘10, pages 91–100,

New York, New York, USA, 2010. [7]. R. Bell, Y. Koren, and C. Volinsky. Modeling relationships at

multiple scales to improve accuracy of large recommender

systems. In Proc. of KDD ‘07, pages 95–104, San Jose, California, USA, 2007.

[8]. M. Jahrer, A. T¨oscher, and R. Legenstein. Combining predictions

for accurate recommender systems. In Proc. of KDD ‘10, pages

693–702, Washington, DC, USA, 2010. [9]. pertz shoval, Veronica Maidel, Brancha shapira, ―international

journal of information theories and applications", Vol.15.pp. 303-

314,2008. [10]. Novák, V., Perfilieva, I. and Močkoř, J. (1999) Mathematical

principles of fuzzy logic Dodrecht: Kluwer Academic. ISBN 0-

7923-8595-0 [11]. "Fuzzy Logic". Stanford Encyclopedia of Philosophy. Bryant

University. 2006-07-23. Retrieved 2008-09-30.

[12]. Zadeh, L.A. (1965). "Fuzzy sets". Information and Control. 8 (3): 338–353. doi:10.1016/s0019-9958(65)90241-x.

[13]. P. Brusilovsky, A. Kobsa, and W. Nejdl (Eds.): The Adaptive

Web, LNCS 4321, pp. 291 – 324, 2007. © Springer-Verlag Berlin Heidelberg 200

[14]. Research-Paper Recommender Systems: A Literature Survey

[15]. Bela Gipp, Jöran Beel, and Christian Hentschel. Scienstein: A Research Paper Recommender System. In

[16]. Proceedings of the International Conference on Emerging Trends

in Computing (ICETiC‘09), pages 309–315,

[17]. Virudhunagar (India), January 2009. Kamaraj College of

Engineering and Technology India, IEEE

[18]. https://en.wikipedia.org/wiki/Markov_chain [19]. http://www.corporatetech.in/admin/ieee/A%20Fuzzy%20Preferen

ce%20Tree-Based%20Recommender.pdf

https://en.wikipedia.org/wiki/Markov_chain [20]. http://setosa.io/ev/markov-chains/

[21]. https://en.wikipedia.org/wiki/Fuzzy_logic

[22]. http://whatis.techtarget.com/definition/fuzzy-logic [23]. http://in.mathworks.com/help/fuzzy/what-is-fuzzy-

logic.html?requestedDomain=in.mathworks.com

[24]. http://scholarworks.sjsu.edu/cgi/viewcontent.cgi?article=1096&context=etd_projects

[25]. https://www.ismll.uni-

hildesheim.de/pub/pdfs/RendleFreudenthaler2010-FPMC.pdf [26]. http://ijariie.com/AdminUploadPdf/Survey_Paper_on_Fuzzy_Logi

c_Based_Recommended_System_ijariie3553.pdf [27]. http://www.jmlr.org/papers/volume6/shani05a/shani05a.pdf

[28]. Research paper on FTM: Fuzzy Theoretic Model